With development and popularity of the Internet industry and a digital television, a digital video has inadvertently entered our daily life, and has been deeply integrated with various daily businesses, creating an inextricable link. Video compression is an indispensable technology in digital video application. The video compression refers to that redundant information in digital video data is removed by using a data compression technology, so as to reduce a data amount required to represent an original video, and facilitate transmission and storage of video data.
In a conventional video compression technology, a motion vector list of a current prediction block generally needs to be constructed. A conventional method for constructing a motion vector list of a current prediction block is as follows: firstly, a motion information candidate list (Candidate List) that is based on a spatially neighboring block and a time-domain corresponding block and that is of the current prediction block is obtained; after that, sub motion information candidates that are of multiple sub blocks of the current prediction block and that are based on inter-view corresponding blocks of the multiple sub blocks of the current prediction block are obtained, and a preferred sub motion information candidate is selected from the sub motion information candidates of the multiple sub blocks; and further, one candidate is selected from the motion information candidate list, and motion information of the candidate and information about a location of the candidate in the list are obtained; when the candidate has motion information, it is determined whether the motion information of the candidate is the same as motion information of the preferred sub motion information candidate; if they are different, the preferred sub motion information candidate is inserted into the motion information candidate list, or if they are the same, insertion is skipped; or when the candidate has no motion information, another candidate needs to be selected, and motion information of the another candidate and information about a location of the another candidate in the list are obtained; when the candidate has motion information, it is determined whether the motion information of the candidate is the same as motion information of the preferred sub motion information candidate, and then it is determined, according to a result of the determining, whether to insert the preferred sub motion information candidate into the motion information candidate list.
In a process of constructing a motion vector list, a candidate needs to be selected from the motion information candidate list, and motion information of the candidate and information about a location of the candidate in the list need to be obtained; when the candidate has motion information, it is determined whether the motion information of the candidate is the same as motion information of the preferred sub motion information candidate, and then it is determined, according to a result of the determining, whether to insert the preferred sub motion information candidate into the motion information candidate list; or when the candidate has no motion information, another candidate needs to be selected from the motion information candidate list for performing the foregoing operation, therefore, the foregoing method for constructing a motion vector list has relatively high complexity, and takes a relatively long time.